Method for intelligently adjusting I/O access load of data center

Abstract

The invention relates to the technical field of communication, in particular to a method for intelligently adjusting an I / O access load of a data center. The method is applied to the data center, andcomprises the following steps of S1, judging whether the data center is in a high access amount time period when receiving a data access request, and if so, turning to step S2; if not, processing thedata access request according to a normal processing mode; S2, judging whether the data access request is of a preset first access type or not, if yes, processing the data access request in a high access amount time period at present, and if not, processing the data access request after delaying to the end of the high access amount time period. The method is beneficial for post-processing part ofaccess to relieve the operation pressure of the memory network when the memory network of the data center is busy in activity, and reducing the congestion condition of a high access time period, so that the processing speed can be accelerated, the efficiency can be improved, and meanwhile, the power consumption can be reduced.

Description

technical field [0001] The invention relates to the field of communication technology, in particular to a method for intelligently adjusting the I / O access load of a data center. Background technique [0002] The Internet Data Center (Internet Data Center) is a common form of hosting cloud computing, usually deploying many servers, providing users with running software or running services inside the user's computer under the unified server. The amount of I / O reads and writes of the storage network of the data center at the same time has a certain upper limit. If the number of user visits suddenly increases in a short period of time, it will exceed the limit of the I / O processing load of the data center storage network. It is expensive to upgrade and replace equipment in the data center, and it is impossible to upgrade equipment frequently, but the daily user visits are dynamically changing, and there will always be situations of visit congestion and visit idleness. The acce...

AI Technical Summary

Problems solved by technology

It is expensive to upgrade and replace equipment in the data center, and it is impossible to upgrade equipment frequently, but the daily user visits are dynamically changing, and there will always be situations of access congestion and idle access

The access congestion caused by the increase in the number of visits will make the I / O load of the data center storage network run at a high load, requiring a large number of read and write operations for each visit at the same time. The storage network cannot transmit a large amount of data at the same time, and the high-load read and write data The operation will cause the storage network I / O to be congested, the processing speed will be slower, and the processing time will be longer, resulting in more energy consumption, lower efficiency, and unable to complete user access in time

During this high-traffic time period, the user experience is greatly affected

Images